272 questions
- Bountied 0
- Unanswered
- Frequent
- Score
- Trending
- Week
- Month
- Unanswered (my tags)
-3
votes
0
answers
24
views
Raft Consensus: Client requests when a leader is not elected [closed]
I was reading into Raft consensus algorithm where the client requests are served by the leader node. In case a new election term begins, there is a brief period when the leader is not elected. We also ...
2
votes
1
answer
61
views
RPC Call cannot find relevant method in Go in Raft Consensus
I am working on implementing Raft Consensus in Go, and there is an RPC handler that is not registering for some reason. I have cross checked my method signature and it matches the Go format for RPC ...
0
votes
0
answers
38
views
Proving the Log Matching property in Raft consensus protocol
I have been trying to understand the Raft protocol for quite some time now. One thing that has always stumped me is the proof of the Log Matching property. One of my concerns is that the proof in the ...
6
votes
1
answer
139
views
How does CockroachDB handle partitions between the leaseholder and Raft leader of a range?
While CockroachDB attempts to ensure that the leaseholder and Raft leader are the same node, this is not enforced (well aside from the introduction of leader leases). In the event that the Raft leader ...
0
votes
1
answer
39
views
is Lost update possible with RAFT?
Lets say We have a cluster of 5 nodes and A is the leader. Following sequence of events take place:
A sends the replicate change request in parallel to all the followers.
Only B could receive the ...
-1
votes
1
answer
187
views
Does Raft follower store leader ID? If not, how does it redirect requests to leader?
Referring to this table depicted in the Raft paper, I did not find where do followers memorize the leader in any form such as identifier, physical addr, etc. Instead, I only find the leader ID in ...
-1
votes
1
answer
51
views
Must the number of orderer node in smartBFT be at least 4?
I'm trying to understand the smartBFT and RAFT consensus protocols in the Hyperledger Fabric. As you know, starting from Fabric v3.0, smartBFT can be used as an ordered type. I checked the official ...
1
vote
1
answer
105
views
How can I back up my database to io.Writer?
I access an SQLite database via mattn/go-sqlite3. I want to perform a live snapshot of that database. The backup must be saved to an io.Writer instance. Specifically the SaveSnapshot() function for a ...
2
votes
1
answer
149
views
What data is stored in the log compaction snapshot of a Raft-based distributed file system?
I'm working on a Raft implementation as part of my distributed file system and I've run into a problem with the log compaction process.
Accurding to the official Raft paper, when a log reaches a ...
1
vote
4
answers
188
views
How does client handle failures in RAFT-replicated datastores? [closed]
Consider a database like cockroachDB that uses RAFT protocol for replicating data to a replica group owning a partition of the data. How does a client handle a request that fails in such DBs? Because, ...
0
votes
1
answer
58
views
Is there a scenario where the client's data is lost, even when it gets success acknowledgement?
In RAFT, assume we have 5 nodes a,b,c,d & e and that "a" is the leader.
A client connected to "a", sent a write (doing non-idempotent operation OP1) request to "a". &...
2
votes
2
answers
230
views
Client request failure in raft
Imagine a 3 node raft cluster. Each node is in sync has log [1,2,3] and entry 3 is committed by the leader.
Now leader receives an entry 4 but fails to commit it because of unreliable network and ...
0
votes
1
answer
91
views
When a Raft leader with uncommitted changes is going to make the commit?
I was playing with https://raft.github.io/ with this situation, but have a big doubt about what is going on.
After a leader is elected, I make a request to it.
The leader send the append entry with ...
0
votes
1
answer
235
views
Raft algorithm three nodes scenario
When a cluster has three nodes, then one of them is down. According to rule, it can tolerate one node fail. But there are two node left, one is leader, the other is follower, how can leader get ...
0
votes
1
answer
272
views
Raft consensus with a shared log: good or bad idea?
Raft consensus protocol requires nodes to have a replicated log, and all the implementations that I'm aware of require that every node has a durable local storage to keep the log. In cloud-native ...